PARROT 323 of M. Blanchard (Comptes Rendus, xliii. 1097-1100 and xliv. 518-521) has not been regarded as successful, and it cannot be affirmed positively that the latest arrange ment of the Psittaci is really imich more natural than that planned by Buffon one hundred and twenty years ago. He was of course unaware of the existence of some of the most remarkable forms of the group, in particular of Strigops and Nestor; but he began by making two great divisions of those that he did know, separating the Parrots of the Old World from the Parrots of the New, and subdividing each of these divisions into various sections somewhat in accordance with the names they had received in popular language a practice he fol lowed on many other occasions, for it seems to have been with him a belief that there is more truth in the discrimi nation of the unlearned than the scientific are apt to allow. The result is that he produced a plan which is comparatively simple and certainly practical, while as just stated it cannot be confidently declared to be unnatural. However, not to go so far back as twenty years, in 1867-68 Dr Finsch published at Leyden an elaborate monograph of the Parrots, 1 regarding them as a Family, in which he admitted 26 genera, forming 5 Subfamilies: (1) that composed of Strigops (KA- KAPO, ut sup?-.) only; (2) that containing the crested forms or Cockatoos ; (3) one which he named Sittadnx, compris ing all the long-tailed species a somewhat heterogeneous assemblage, made up of MACAWS (vol. xv. p. 130) and what arc commonly known as Parakeets ; (4) the Parrots proper with short tails ; and (5) the so-called " brush- tongued " Parrots, consisting of the LORIES (vol. xv. p. 7) and NESTORS (ut sup.). Except in the characters of the last group he recognized none that were not external, and that fact is sufficient to cast suspicion on his scheme being natural. In 187-4 the late Prof. Garrod communicated to the Zoological Society the results of his dissection of examples of 82 species of Parrots, which had lived in its gardens, and these results were published in its Proceedings for that year (pp. 586-598, pis. 70, 71). The principal points to which he attended were the arrangement of the carotid artery, and the presence or absence of an ambiens muscle, an oil-gland, and a furcula ; but except as regards the last character he unfortunately almost wholly neglected the rest of the skeleton, looking upon such osteological features as the formation of an orbital ring and peculiarities of the atlas as " of minor importance " an estimate to which nearly every anatomist will demur ; for, though undoubtedly the characters afforded by blood-vessels and muscles are useful in default of osteological characters, it is obvious that these last, drawn from the very framework of any vertebrate s structure, cannot be inferior in value to the former. Indeed the investigations of Prof. A. Milne-Edwards (Ann. Sc. Nat. Zoologie, ser. 5, vi. pp. 91-111 ; viii. pp. 145-156) on the bones of the head in various Psittacine forms make it clear that these alone present features of much significance, and if his investigations had not been carried on for a special object, but had been extended to other parts of the skeleton, there is little doubt that they would have removed some of the greatest difficulties. The one osteological character to which Garrod trusted, namely, the condition of the furcula, cannot be said to contribute much towards a safe basis of classification. That it is wholly absent in some genera of Parrots had long been known, but its imperfect ossification, it appears, is not attended in some cases by any diminution of volant powers, which tends to shew that it is an unimportant character, an inference confirmed by the fact that it is found wanting in genera placed geographically so far apart that the loss must have had in some of them an independent origin. Summarily expressed, Garrod s scheme was to divide the Parrots into two Families, Palseomithidx and Psittacidee, assigning to the former three Subfamilies Palfeornithina?, Cacntuinsc, and Stringopinx, and to the latter four, Anna?, Pyrrhiirinse, PlatycercinsR, and Chrysotinse. That each of these sections, except the Ccicatuinse, is artificial any regard to osteology would shew, and it would be useless here to further criticize his method, except to say that its greatest merit is that, as before mentioned (LovE-BiRT), vol. xv. p. 28), he gave sufficient reasons for distinguishing between the genera Agapornis and Psittacula. In the Journal fur Ornithologie for 1881 Dr Reichenow published a Conspectus Psittacorum, founded, as several others 2 have been, on external characters only. He makes 9 Families of the group, and recognizes 45 genera, and 442 species, besides subspecies. His group ing is generally very different from Garrod s, but displays as much artificiality ; for instance, Nestor is referred to the Family which is otherwise composed of the Cockatoos. Still more recently we have the arrangement followed by Mr Sclater in the List of those exhibited of late years in the gardens of the Zoological Society, and published in 1883. This is more in accordance with the views that the present writer is inclined to hold, and these views may here, though with much diffidence, be stated. First there is Strigops, which must stand alone, unless, as before hinted (vol. xiii. p. 826), Geopsittacus and Pezopoms may have to be placed with it in a Family Strigopidse. Next Nestor, from its osteological peculiarities, seems to form a very separate type, and represents a second Family Nestoridm. These two Families being removed, all the Parrots that remain will be found to have a great resemblance among them selves, and perhaps it is impossible justifiably to establish any more Families. For the present at any rate it would seem advisable to keep them in a single Family Psittacida?, but there can be no objection to separating them into several Subfamilies. The Cockatoos, for instance, can be without much difficulty defined, and may stand as Cacatuinse, and then the brush-tongued Lories as Loriinie, after which the Macaws, Arinse including possibly Conurus and its allies. Platycercus and its neighbours may form another section, and the same with Paleeornis ; but for the rest there is not yet material for arriving at any determination, though Chrysotis and Psittacus seem to furnish two different types, to the former of which Psittacula appears to bear much the same relation as Agapornis does to the latter. Amongst the genera Chrysotis, Pal&ornis, and Psittacus are probably to be found the most highly organized forms, and it is these birds in which the faculty of so-called " speech " reaches its maximum development. But too much import ance must not be assigned to that fact ; since, while Psittacus erithacus the well-known Grey Parrot with a red tail is the most accomplished spokesman of the whole group, it is fairly approached by some species of Chrysotis usually styled Amazons and yet its congener P. timneh is not known to be at all loquacious. 3 Considering the abundance of Parrots both as species and individuals, and their wide extent over the globe, it is surprising how little is known of their habits in a wild state. Even the species with which Englishmen and their descendants have been more in contact than any other has an almost unwritten history, compared with that of many other birds; and, seeing how it is oppressed by and yielding to man s occupation of its ancient haunts, the 2 Such, for instance, as Kuhl s treatise with the same title, which appeared in 1820, and Wagler s Monograpliia Fsittacormn, published in 1832 both good of their kind and time. 3 In connexion with the "speaking" of Parrots, one of the most curious circumstances is that recorded by Humboldt, who in South America met with a venerable bird which remained the sole possessor of a literally dead language, the whole tribe of Indians, Atures by name, who alone had spoken it having become extinct.
